5 months ago
London, United KingdomSenior
Responsibilities
- Run experiments self-hosting models across cloud instances and on-premises hardware configurations.
- Develop reproducible and measurable deployment patterns optimized for latency, throughput, and cost.
- Improve orchestration and routing software, including caching, request scheduling, batching, and resource allocation.
- Integrate accelerator support, inference engine features, and infrastructure upgrades while identifying bottlenecks and required capabilities.
- Build and maintain benchmarking harnesses, regression suites, and performance dashboards.
- Translate performance data into actionable feedback and guide production-readiness and stack-optimization decisions.
Requirements
- Experience deploying and benchmarking large-model inference in production or production-equivalent environments.
- Familiarity with multi-node GPU deployments and associated networking and communication stacks.
- Strong end-to-end performance characterization skills, including isolating bottlenecks in networks, runtimes, memory subsystems, or models.
- Familiarity with serving frameworks such as Dynamo, Triton Inference Server, or similar orchestration layers.
- Clear communication skills and the ability to turn performance data into prioritized feedback.
- A disciplined and systematic approach to reproducible deployment, measurement methodology, and controlled comparisons.
Benefits
- Competitive salary determined by skills and experience.
- Equity and ownership.
- Private healthcare.
- Visa sponsorship and relocation benefits.
- In-person work at the London office with provided tools, workspace, and setup.
Categories
About Callosum
Callosum builds AI infrastructure that unifies heterogeneous compute—software and systems to orchestrate diverse accelerators and the physical clusters they run on—for organizations training and serving models. The privately held company was founded in 2025 and is headquartered in London. It focuses on designing, deploying, and operating mixed-hardware compute fleets across sites and vendors, and raised a seed round in 2026.